The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 15 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Masad: ChatGPT has triggered psychotic breaks by reinforcing delusions
“There's high-profile cases where people have psychotic breaks because they have, like, delusions of grandeur and things like that, and ChatGPT is, like, reinforcing that, and that's, like, really bad.”
Gavet: Startups attempting to compete with ChatGPT will probably fail
“And frankly, if you're an entrepreneur and your dream is to compete with ChatGPT, you're probably not going to succeed because of the compute power and the data sets required to even remotely have a shot at competing. So the only businesses that might compete …”
Suleyman: Google's LaMDA Was 'ChatGPT Way Before ChatGPT'
“So in many ways, it was ChatGPT way before ChatGPT, and we were completely blown away with how good your seventh and 10th turn of conversation was with the model, because it obviously had the prior turns of interaction in its working memory.”
Hoffman: Multiple AI agents will coexist rather than one dominant winner
“My view is we're going to actually have a world of a number of different agents. I don't think there's going to be one agent to rule them all”
Masad: Kids will learn programming by osmosis using Replit and ChatGPT
“And I promise you, your kids will learn programming by osmosis. Like, They can just, like, see, they're gonna get curious at some point, they run into a bug, they're gonna go around and ask you at GPT, and that's how I think the organic learning process happen…”
Hoffman: AI hallucinations can still accelerate human productivity
“And so even in this kind of hallucination case because of the research, it was actually still an accelerant. It just is an accelerant in a way that kind of proves the point from my earlier book, Impromptu, and also part of what I'm trying to say in Superagency…”
Taylor: Conversational interaction is the most ergonomic software interface
“My intuition is that with the advent of large language models and ChatGPT being that inflection point with large language models, talking to software is perhaps the most ergonomic way we can interact with software because you don't need an instruction manual t…”
Lakhani: Despite widespread trial, only 10% of professionals use generative AI daily
“So, like, 80%, 90% say they've used it. 75% to 70% say that it's going to change their job, their career, whatever. Then I ask, how many of you use these tools every day in your professional life? 10%. 10% say yes. So what I see is, there's a massive knowing d…”
Kawasaki: Leading AI models lack distinct brand positioning and differentiation
“Like today, if I had to explain to someone when to use ChatGPT, when to use Bard, when to use Perplexity, when to use Claude, when to use all this, you know, I don't know if I could, I can't. I mean, I use all fives kind of at random, and it's like, in the car…”
Gelsinger: ChatGPT-scale AI breakthroughs will emerge every two years for a decade
“But the next two years, we're going to have the next chat GPT. And then two years later, the next one and the next one and the next one. You know, we see a decade or more of continued evolution.”
Hoffman says he was unaware OpenAI was launching ChatGPT
“I didn't know they were going to launch ChatGPT when they were doing it. That was outside my awareness and focus.”
Lakhani: ChatGPT generates willingness-to-pay estimates that match real consumer data
“My colleague Ayelet Israeli has a consumer intelligence lab at our institute, and she's, she has shown that you could do consumer research with ChatGPT. You can do market research. You can get willingness to pay estimates that match real consumer data”
Bret Taylor: AI hallucinations violate 30 years of deterministic computing expectations
“Because we have been trained for the past 30 years to think of Computers as these extremely powerful rules engines. A correctly running computer was entirely deterministic, right? A correctly running computer could look up facts. When you search something in G…”
Gustafsson: Generative AI creates linguistically complex and convincing phishing emails
“We're seeing phishing emails being crafted using these tools. The linguistic complexity of those emails has got far broader as it becomes much easier for people to communicate in a way that feels much more natural and human.”